Spinal Cord Atrophy and Reorganization of Motoneuron Connections Following Long- Standing Limb Loss in Primates  Carolyn W.-H. Wu, Jon H. Kaas  Neuron  Volume 28, Issue 3, Pages 967-978 (December 2000) DOI: 10.1016/S0896-6273(00)00167-7

Figure 1 Distribution of Labeled Motoneurons in the Lateral Motoneuron Pools in the C6 and C8 Segments of the Spinal Cord in Normal Squirrel Monkeys Photomicrographs are taken from spinal cord cross-sections. Motoneurons were retrogradely labeled following injections of CTB into (A) distal forelimb muscles or (B) shoulder muscles of a normal monkey. The level in the spinal cord is shown at the upper-right corner for each section. Within the lateral motoneuron groups (i.e., Rexed's lamina IX), the ventromedial portion projects to the proximal limb, whereas the dorsolateral portion projects to distal limb. Note that the distribution of labeled neurons innervating the shoulder muscles are complimentary to those innervating the distal forelimb muscles. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 2 Photomicrographs of a Series of Spinal Cord Cross-Sections from Forelimb-Amputated Galago and Squirrel Monkey Showing the Distribution of Retrogradely Labeled Spinal Motoneurons The lower magnification views of the sections are in the middle. The lateral motoneuron pools on both sides of the spinal cord are shown in the enlarged views. (A) Sections from a forelimb-amputed galago (case 97-100) prepared for CTB exhibit the well-defined somata and dendritic arbors of individually labeled neurons. Following CTB injections into muscles of the intact hand, labeled neurons on the intact side of the spinal cord were found in the most dorsolateral extreme of the lateral motoneuron pools. Following CTB injections into the limb stump, labeled neurons on the amputated side were found in the dorsolateral portion of the lateral motoneuron pools controlling the distal forelimb. Note that labeled dorsolateral neurons on the amputated side are small and have simpler processes than those on the intact side. Also note the apparent reduction in volume of both ventral and dorsal horns on the amputated side of the spinal cord. (B) Dark-field photomicrographs from a forelimb-amputated squirrel monkey (case 98-61) showing retrogradely labeled neurons (arrowhead) following FB injections into the distal forelimb and limb stump muscles. Injections into the limb stump labeled lateral motoneuron pools that would normally innervate the distal forelimb muscles, as identified by the location of labeled neurons after forelimb injections on the intact side. Similar results were obtained in case 98-64. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 6 Summary Diagram of Reorganization at Different Levels of the Motor System after Long-Term Limb Loss (Top) Pie charts showing the percentage of intracortical microstimulation sites for different body movements in the forelimb cortex of M1 in a forelimb-amputated squirrel monkey (case 98-61; Wu and Kaas 1999). In M1 ipsilateral to the missing limb (i.e., intact M1), evoked movements predominantly involve forelimb and shoulder muscles, with a small percentage of sites devoted to nonlimb movements involving the face and trunk muscles. By contrast, in M1 contralateral to the missing limb (i.e., deefferented M1), there is a significant increase in the percentage of sites devoted to shoulder or stump and nonlimb movements. (Bottom) Schematic drawings of the somatotopic organization of various motoneuron pools of the spinal cord innervating different muscle groups. The segmental distribution of motoneurons controlling distal forelimb, proximal forelimb, shoulder, or upper trunk muscles on the intact side are illustrated on the left. On the right, the mediolateral and dorsoventral distributions at different segmental levels of the motoneurons on the amputated side are compared to those on the intact side. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 3 Photomicrographs of a Series of Spinal Cord Cross-Sections from a Forelimb-Amputated Squirrel Monkey Photomicrographs from case 98-64 show the distribution of retrogradely labeled spinal motoneurons in sections processed for CTB. Following CTB injections into muscles of the forelimb and the shoulder on the intact side, labeled neurons were found in both lateral and medial motoneuron pools throughout C5–T2 segments (left). Injections of CTB into the shoulder muscles on the amputated side resulted in a similar extent of label as on the intact side (right). This indicates that some motoneurons formerly controlling the amputated limb have changed their muscle targets to innervate the shoulder muscles. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 4 Reorganization of Motoneurons in Hindlimb-Amputated Animals (A) Photomicrographs of a series of spinal cord cross-sections showing the distribution of retrogradely labeled motoneurons in the ventral horn of normal galagos after CTB injections into the muscles of the left upper leg and right foot. Motoneurons innervating the muscles of the feet were located in the lateral extreme of the lateral motoneuron pools at L6–S2 levels, whereas motoneurons innervating upper leg muscles were located in the ventromedial portion of the lateral motoneuron pool at L4–S1 levels. (B) Photomicrographs of a series of spinal cord cross-sections from a leg-amputated galago (case 97-134) showing the distribution of retrogradely labeled spinal motoneurons in sections processed for CTB. Following bilateral and symmetrical injections of CTB into muscles of the hip, labeled neurons on the intact and amputated sides exhibited similar distributions, but more extensive labeling was found on the amputated side at lower levels, including dorsolateral portions of the lateral motoneuron pools that normally innervate distal hindlimb muscles. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 5 Frequency Distribution of Cross-Sectional Areas for the Population of Dorsolateral Motoneurons on the Amputated Side and Intact Side in Animals with an Amputated Limb Black bars indicate the amputated side and gray bars indicate the intact side. Motoneurons were sampled from the C7–T2 levels in the forelimb-amputated galago (left) as well as L6–S2 levels of the hindlimb amputated squirrel monkey (right). Note the bimodal distribution of motoneurons on the intact side, with a reduction in numbers between the presumably smaller γ motoneurons and the larger α motoneurons. In contrast, on the amputated side, there is an obvious decrease in the number of motoneurons within the α motoneuron size range. Note that the total numbers of motoneurons sampled from the two sides are not significantly different within the same animals. Bin size is 100 μm2. The last bin represents values equal to or greater than the value indicated for that bin. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)

Figure 7 Photomicrographs of Cross-Sections from the S1 Segment of the Spinal Cord in Hindlimb Amputees Showing Changes in Motoneuron Morphology Photomicrographs of cross-sections reacted for Nissl are shown in (A), for SMI-32 in (B), and for AChE in (C). Pairs of images are enlarged from intact (left) and amputated (right) sides of the same spinal sections. The general decrease in the size of dorsolateral nuclei on the amputated side can be clearly seen in these sections. Also note that the somata and primary dendrites of motoneurons on the amputated side are smaller than those on the intact side. Scale bar, 100 μm. Neuron 2000 28, 967-978DOI: (10.1016/S0896-6273(00)00167-7)